sign in sign up
<<
Home , Kolumbo, Submarine eruption, Volcanic arc

OceTHE OFFICIALa MAGAZINEnog OF THE OCEANOGRAPHYra SOCIETYphy

CITATION Embley, R.W., Y. Tamura, S.G. Merle, T. Sato, O. Ishizuka, W.W. Chadwick Jr., D.A. Wiens, P. , and R.J. Stern. 2014. Eruption of South Sarigan , Northern Mariana : Insights into hazards from submarine volcanic eruptions. 27(2):24–31, http://dx.doi.org/10.5670/oceanog.2014.37.

DOI http://dx.doi.org/10.5670/oceanog.2014.37

COPYRIGHT This article has been published in Oceanography, Volume 27, Number 2, a quarterly journal of The Oceanography Society. Copyright 2014 by The Oceanography Society. All rights reserved.

USAGE Permission is granted to copy this article for use in teaching and research. Republication, systematic reproduction, or collective redistribution of any portion of this article by photocopy machine, reposting, or other means is permitted only with the approval of The Oceanography Society. Send all correspondence to: [email protected] or The Oceanography Society, PO Box 1931, Rockville, MD 20849-1931, USA.

DOWNLOADED FROM HTTP://WWW.TOS.ORG/OCEANOGRAPHY SPECIAL ISSUE ON UNDERSEA NATURAL HAZARDS

Eruption of South Sarigan Seamount, Northern Insights into Hazards from Submarine Volcanic Eruptions

BY ROBERT W. EMBLEY, YOSHIHIKO TAMURA, SUSAN G. MERLE, TOMOKI SATO, OSAMU ISHIZUKA, WILLIAM W. CHADWICK JR., DOUGLAS A. WIENS, PATRICK SHORE, AND ROBERT J. STERN

ABSTRACT. The eruption of South Sarigan Seamount in the southern Mariana arc activity. Seismic body waves and in May 2010 is a reminder of how little we know about the hazards associated with T-phases (waterborne acoustic waves) submarine explosive eruptions or how to predict these types of eruptions. Monitored produced by the eruptive events were by local seismometers and distant hydrophones, the eruption from ~ 200 m water recorded on seismometers on nearby depth produced a gas and ash plume that breached the surface and rose ~ 12 km islands (Searcy, 2013) and on submarine into the atmosphere. This is one of the first instances for which a wide range of pre- hydrophones and seismometers located and post-eruption observations allow characterization of such an event on a shallow 2,260 km to the east of South Sarigan submarine . Comparison of bathymetric surveys before and after Seamount. The mid-Pacific instruments the eruptions of the South Sarigan Seamount reveals the eruption produced a 350 m near Wake are operated by the diameter crater, deeply breached on the west side, and a broad apron downslope with International Monitoring System (IMS) deposits > 50 m thick. The breached summit crater formed within a pre-eruption of the Comprehensive Nuclear Test dome-shaped summit composed of . Dives with the Japan Agency for Ban Treaty (CTBT) (Green et al., 2013). Marine-Earth Science and Technology Hyper-Dolphin remotely operated vehicle Precursory volcano-tectonic sampled the wall of the crater and the downslope deposits, which consist of andesite began in early April 2010 and continued blocks lying on pumiceous gravel and . Chemical analyses show that the intermittently through April and early andesite is probably juvenile material from the eruption. The unexpected May, with a significant increase in size eruption of this seamount, one of many poorly studied shallow of and number of swarms beginning on comparable size along the Mariana and other volcanic arcs, underscores our lack of May 11 (Searcy, 2013). Hydroacoustic understanding of submarine hazards associated with submarine . phases, interpreted as eruptive activity (Green et al., 2013; Searcy, 2013), began AN UNUSUAL ERUPTION an underwater eruption occurred on on May 27, reaching a peak during ON THE MARIANA ARC South Sarigan Seamount, a poorly a three-hour period early on May 29 An unexpected submarine volcanic surveyed lying with near-continuous volcanic tremor. eruption on the southern Mariana arc between Sarigan and Anatahan Islands Also during this time span, patches of (Figure 1a,b) in 2010 provides insights in the Commonwealth of the Northern discolored water were observed on the into the volcanic processes of submarine Mariana Islands (Figure 1a,b). The surface (McGimsey et al., 2010) eruptions and the hazards these hidden eruption was unusual because this and infrasound events (atmospheric volcanoes pose. On May 28–29, 2010, seamount had no known historical acoustic signals) were recorded on a

24 Oceanography | Vol. 27, No. 2 143°E 144°E 145°E 144°E 145°E 146°E 143°E 144°E 145°E 144°E 145°E 146°E 143°E 144°EDashed lines indicate145°E 142°E 144° M144°E146° 145°E 146°E 143°E 144°E 145°E a144°E 145°E 146°E a summit depths <500m a ri b Dashed lines indicate 142°E 144° M a 146° Dashed lines indicate 142°E 144° M n146° Guguan summit depths <500m a M ar a a a summit depths <500m a ar i T b b a Nikko summitComposite depths Cone <500m a riaa r b iann e Guguan naa n Guguan 22° a 143°E 144°E Composite 145°ECone 144°ET T c 145°E 146°E NikkoNikko BasalticComposite Cone Cone Tr re h Nikko re n 23°N

22° n

22° n c

22° c BackarcDashedBasalticBasaltic linesAxis Cone Cone indicate 142°E 144° M 146° chh 17°N Basaltic Cone a h Zealandia (s) 23°N 23°N summit depths <500m a 23°N a ri b

a 17°N Backarc Axis 17°N Backarc Axis n 17°N MarianaBackarc Islands Axis a W Sarigan GuguanZealandiaZealandia (s) (s) M T Sarigan Nikko Composite Cone Maug r 20° e MarianaMariana Islands Islands a WW Sarigan Sarigan Mariana Islands MM n W Sarigan SariganSarigan Syoyo 22° c Sarigan Maug ri 20° Maug h South Basaltic Cone 20° 20° aaa 23°N SyoyoSyoyo Syoyo a ri South >5000 0 17°N DepthBackarc (m) Axis riri ZealandiaSouth (s)SouthSarigan n Anatahan a >5000>5000 DepthDepth (m) (m) 00 a a SariganSariganSeamount n a Anatahan n Anatahan Mariana Islands M n W SariganAnatahan Sarigan

Seamount a Seamount Maug a Seamount a

20° Pagan A

18° a

Syoyo

Pagan Pagan A ri A 18° South Pagan r 18° A

Fukujin 18° c

a r Fukujin >5000 Depth (m) 0 r Sarigan

b r

22°N

Fukujin c c

Fukujin n c Anatahan b c 22°N b Diamante (s) 22°N b Seamount 16°N a 22°N Diamante (s)

Diamante (s) 16°N Diamante (s) 16°N

S SariganPagan 16°N A 18° S Sarigan

Fukujin r

16° c c 16°

16° b 22°N

16° Diamante (s) Kasuga-2Kasuga-2 RubyRuby 16°N Kasuga-2 EifukuEifuku Saipan Ruby Eifuku S SariganSaipan Saipan W Saipan DaikokuDaikoku 16° WW Saipan Saipan Kasuga-2 Daikoku W Saipan Ruby Eifuku Saipan Saipan 14°N 14°N h SaipanSaipan 14°N c h Daikoku nc c EsmeraldaW Saipan Saipan 14°N Guam e n h Esmeralda Guam rre nc Km 100 T ren Esmeralda

21°N Guam 21°N Km 100 Tre Tinian 21°N 15°N Km14°N 100 T Saipan h Tinian 15°N

21°N c Esmeralda Tinian 15°N Guam en Farallon de 15°N Farallon deKm 100 Tr 21°N FarallonFarallonPajaros de de W TinianTinian Pajaros W Tinian 15°N PajarosFarallon de WW T Tinianinian Ahyi Makhahnas PajarosAhyi W Tinian Makhahnas AhyiAhyi W Rota Makhahnas Ahyi Makhahnas W Rota Supply Maug W Rota Maug Rota SupplyReef MaugMaug

20°N Supply 20°N Supply

Rota 14°N Rota 14°N ReefReefReef Rota 20°N 20°N 20°N 14°N 14°N 14°N Asuncion TM proj. CM 144°E TM proj. CM 144°E Asuncion TM proj. CM 145°E TM proj. CM 144°E AsuncionAsuncionCheref Guam TM proj. CM 145°E TM00 proj. CMkm 144°E 100100 0 TM proj.km CM 145°E100 TM proj. CM 144°Ekm Cheref Guam TMkm proj. CM 145°E 0 km 100 CherefCheref GuamGuam 0 km 100 0 0 km km 100100 0 0 kmkm 100100 c 175 d c 175500 d 500 c c 175175 dd 1000500 1000500

1500 Depth (m) Depth (m) 15001000 Depth (m) 1000

200020001500 Depth (m) 1500 Depth (m) 500500 m m mid-imagemid-image2000 2000 P2P2 500 m P2 P2 P1P1 500 m P1P1 mid-image P1 P2 mid-image P2 P1 P2 P1 P2

N N N N e Difference Grids EM122 MV1302a f Difference Grids e MV1302a - TN153 (positive changes) N EM12225 m Bathymetry grid-cell size MV1302a Nf MV1302aMV1302a - TN153 - Combo (positive of TN153 changes) & EW0202 (negative changes)N N 25 m grid-cell size -1400 MV1302a - Combo of TN153 & EW0202 (negative changes) -1400 16°38'N e Difference Grids EM122 Bathymetry MV1302a 16°38'N f 16°38'N 16°38'N 16°38'N DifferenceMV1302a Grids - TN153 (positive changes)+100? EM122 Bathymetry MV1302a 16°38'N e +100? 25 m grid-cell size f MV1302aMV1302a - TN153 - Combo (positive of TN153 changes) & EW0202 (negative changes) -1000 25 m grid-cell size -1400 MV1302a - Combo of TN153 & EW0202 (negative-50 changes)-150 -1000

16°38'N -1400 16°38'N 16°37'N -50-50 -150-150 16°37'N

16°38'N +100? P1 -2000 P1 16°38'N 16°37'N 16°37'N 16°37'N 16°37'N +100?+50 -200 16°37'N P1 -2000-2000 P1 -1000 +50 -50 -200 -1000 +50 -150 -500 -300 -1000 16°36'N 16°36'N 16°37'N +50 to +100 -50 -1000-1000 16°37'N -150 -1500 -500 -300 Polygon constraint P1 -2000 P1 16°37'N 0 16°37'N 16°36'N 16°36'N Polygon constraint 16°36'N +50+50 toto +100+100 for positive volumes -200 16°36'N +50 for negative volumes -1500-1500 -200 Depth P1 -2000 P1 00 Changes Polygon constraint Polygon constraint-200 +50 -1000 forfor positivepositive volumesvolumes -200

16°35'N -500 Depth-100 to - 200 forfor negativenegativeP2 volumesvolumes P2 -300 16°35'N 16°36'N +50 toChanges +100 -1000 16°36'N 0 Km 2 -500 0 Km 2 -1500 -800 -300 16°35'N 16°35'N 16°35'N -100-100Geographic toto -- 200200 proj. P2 P2 16°35'N 16°36'N +50 0to +100 Polygon constraint Polygon constraint Geographic proj. 16°36'N -1500 for positive volumes 145°42'E 145°43'E 145°44'E 145°45'E 145°46'E -200145°47'E 145°42'E00DepthKm 145°43'E22 145°44'E 145°45'E for negative145°46'E volumes 0 Km 2 -800 0 Geographic proj. Polygon constraint Polygon constraint ChangesFigure 1 for positive volumes Geographic proj. Depth for negative volumes -200 16°35'N 145°42'E145°42'E-100 to - 145°43'E200145°43'E 145°44'E145°44'E 145°45'E P2145°46'E 145°42'E 145°43'E 145°44'E 145°45'E 145°46'EP2 145°47'E 16°35'N Changes Figure 1

16°35'N 0 2 -100 toKm - 200 P2 0 Km 2 P2 -800 16°35'N Figure 1. Maps showingGeographic the proj. regional setting and details of the South Sarigan eruptionGeographic site. The proj. inset between the top panels locates the two maps. (a) Northern 0 Km 2 0 Km 2 -800 Seamount Province145°42'E of the145°43'E Mariana arc145°44'E (Bloomer145°45'E et al., 1989).145°46'E Seamounts with145°42'E summit depths145°43'E < 500 m145°44'E have dashed145°45'E outlines. (b)145°46'E Seamounts145°47'E of the Southern Geographic proj. Geographic proj. Seamount ProvinceFigure indicating1 the location of South Sarigan Seamount. (c) Fledermaus © 3D image made from a combination of pre-eruption 2002 and 2003 bathymetry,145°42'E view looking145°43'E east toward145°44'E the summit145°45'E of South145°46'E Sarigan Seamount.145°42'E P1 and P2 145°43'Eare, respectively,145°44'E the eruption145°45'E site and 145°46'Ethe larger summit145°47'E peak that was Figure 1 unsurveyed before 2013. Note P1’s dome shape. (d) Fledermaus © 3D image made from post-eruption 2013 bathymetry, view looking east toward the summit of South Sarigan Seamount. Note the new crater on peak P1 and new bathymetry for peak P2, shown as a shallow flat white area. (e) Depth change map. Positive depth changes were determined by differencing the 2003 (TN153) bathymetry data with the 2013 (MV1302) bathymetry grid. The red polygon constrains the area used to calculate positive volumes. Negative depth changes were determined by differencing a combined grid of 2002 (EW0202) and 2003 (TN153) bathymetry data with the 2013 (MV1302) bathymetry grid. The blue polygon constrains the area used to calculate negative volumes. Contours represent 50 m depth change intervals; note negative depth change numbers at and near P1 where the crater is located and positive depth change contours downslope and west of the crater. (f) Post-eruption 2013 bathymetry map of South Sarigan Seamount with 100 m contours. Oceanography | June 2014 25 station in Palau (Green et al., 2013). After TYPES OF SUBMARINE 1981). Various causal mechanisms a short (~ 3 hour) quiescent period, the ERUPTION HAZARDS have been debated for these , paroxysmal explosive events on May 29 Although most submarine eruptions including submarine collapse, produced the largest peak-to-peak hydro- occur along the mid-ocean ridge in deep surges into the ocean, acoustic signals and another large infra- water and without any impact on society and underwater explosions. The largest event (Green et al., 2013; Searcy, (Rubin et al., 2012), there are significant associated with the paroxysmal 2013). The final phase also generated a numbers of potentially hazardous shal- eruption of Thera was likely caused small tsunami with up to ~ 3 m runup on low submarine volcanoes above oceanic by rapid submarine caldera formation nearby Sarigan Island and 4–5 cm mea- hotspots and along volcanic arcs within (McCoy and Heiken, 2000). The mecha- sured on gauges in Saipan, located zones. When compared nism for the Krakatau tsunami remains about 270 km SSW (McGimsey et al., to mid-ocean ridges, these volcanoes, controversial, although numerical wave 2010). A biology field party on Sarigan particularly those along the volcanic modeling by Nomanbhoy and Satake Island (11 km to the north) heard a loud arcs, are characterized by more explosive (1995) favors an underwater explosive noise and observed an eruption plume eruptions due to their shallower depths, origin. Maeno et al. (2006) modeled rising from the ocean. The eruption higher volatile contents, and more tsunami inundations along the southern plume was tracked by satellite, reaching viscous . Most of the hundreds Kyushu (Japan) from a large sub- ~ 12 km into the atmosphere and trigger- of seamounts in these settings are in marine caldera collapse at offshore Kikai ing a volcanic ash advisory statement for the western Pacific, Mediterranean, and volcano 7,300 years ago. They determined air traffic from the Washington Volcanic Caribbean areas (e.g., Mariana arc shown that the amount of inundation is strongly Ash Advisory Center. The overlying in Figure 1a,b). The large majority of dependent on the rate of caldera collapse. apparently prevented most these submarine volcanoes have poorly Large tsunamis can also be caused by of the mass flux of the eruption from known geologic histories and hazards. very large flank failures on submarine breaching the ocean surface, consistent Historically, the major regional (up to volcanoes or hybrid (subaerial/ with satellite and other observations that hundreds of kilometers) hazard associ- submarine) failures from landslides asso- the atmospheric plume was mostly vapor ated with island and/or submarine erup- ciated with the collapses of island flanks. (McGimsey et al., 2010). This event is tions has been tsunamis (Latter, 1981; We will not address this class of hazard one of the few documented submarine Mastin and Witter, 2000; Paris et al., in this paper because it is not necessarily eruptions originating from a water depth 2013). The large “” gen- directly related to submarine eruptions > 100 m that have breached the sea erated during the paroxysmal eruptions and is covered elsewhere in this special surface (Mastin and Witter, 2000; Kano, of Krakatau and Thera () were issue (see Watt et al., 2014). 2003; Carey et al., 2014). responsible for mass fatalities (Latter, Locally hazardous tsunamis have also been generated from smaller underwater Robert W. Embley ([email protected]) is Senior Research Scientist, National volcanic explosions such as those at Oceanic and Atmospheric Administration, Pacific Marine Environmental Laboratory, Myōjin-shō (a shallow submarine Newport, OR, USA. Yoshihiko Tamura is Principal Scientist, Japan Agency for Marine- volcano on the Izu-Ogasawara arc, south Earth Science and Technology (JAMSTEC), Yokosuka, Japan. Susan G. Merle is Senior of Japan) in the early 1950s (Dietz and Research Assistant, Cooperative Institute for Marine Resources Studies, Oregon State Sheehy, 1954). These types of volcanic University, Newport, OR, USA. Tomoki Sato is Technical Support Staff Member, JAMSTEC, tsunamis have also been reported from Yokosuka, Japan. Osamu Ishizuka is Senior Scientist, Geological Survey of Japan of submarine eruptions near Ritter Island the Advanced Industrial Science and Technology Agency, Tsukuba, Japan, and Visiting (Papua New Guinea) and Kick’em Jenny Senior Scientist, JAMSTEC, Yokosuka, Japan. William W. Chadwick Jr. is Professor, volcano in the Caribbean (Beget, 2000). Cooperative Institute for Marine Resources Studies, Oregon State University, Newport, Submarine eruptions that breach the OR, USA. Douglas A. Wiens is Professor, Department of Earth and Planetary Sciences, sea surface can threaten vessels, air Washington University, St. Louis, MO, USA. Patrick Shore is Computer Specialist/Lecturer, traffic (from the ash cloud), and even Department of Earth and Planetary Sciences, Washington University, St. Louis, MO, USA. nearby land masses. This direct hazard Robert J. Stern is Professor of Geosciences, University of Texas at Dallas, Dallas, TX, USA. is exemplified by the tragic loss of the

26 Oceanography | Vol. 27, No. 2 Japanese hydrographic vessel No. 5 widespread pumice raft and atmospheric 2013). That survey shows a dome-shaped Kaiyo Maru during investigations of plume formed above Havre Seamount summit rising up to ~ 184 m water Myōjin-shō volcano in 1952 (Niino, (Kermadec arc) during its 2012 eruption, depth (Figure 1c). A post-event survey 1952; Dietz and Sheehy, 1954). Such apparently from a source at > 700 m in February 2013 (Simrad EM122 eruptions can also produce large pumice depth (Carey et al., 2014), underscoring multibeam, cruise MV1302) reveals that rafts and sometimes build cones that our poor understanding of the depth a breached crater, 350 m in diameter, and reach into very shallow water and limit below which submarine volcanoes with up to ~ 200 m relief, had formed on construct ephemeral islands that can are no longer a threat at the surface. the northern peak in the interval since be navigational hazards and that often the 2002 survey (P1 on Figure 1d,f). disappear due to wave (Vaughan PRE- AND POST-ERUPTION We are assuming that all the depth et al., 2007; Carey et al., 2014). Rapid BATHYMETRY OF SOUTH changes occurred as the result of the discharge of magmatic gases into the SARIGAN SEAMOUNT May 2010 eruptions because available overlying water column is a potential During the past decade, there has been seismicity records from stations on the submarine volcanic hazard above very a dramatic increase in exploration nearby islands of Sarigan and Anatahan shallow seamounts because sufficiently and research on the intra-oceanic arcs do not show any significant activity in high gas concentrations can cause a (produced by the subduction of one the vicinity of South Sarigan Seamount loss of buoyancy for vessels (similar to oceanic plate beneath another) of the except for the period of 2010 described the hazard caused by blowouts while western Pacific by various groups using above (Cheryl Searcy, Alaska Volcano drilling for hydrocarbons; Hovland and multibeam and side-scan sonars for Observatory, pers. comm., 2014). Gudmestad, 2001). Degassing episodes broad-scale mapping and remotely To quantify the changes that took can occur even without accompanying operated vehicles for in situ studies place during the eruption, we used eruptive activity either suddenly by (Embley et al., 2007, 2008). These efforts the MB-system © to create difference tectonic triggering (Esposito et al., 2006) have only begun to assemble the geologic grids between the 2013 multibeam or from a slow buildup of magmatic history of these hundreds of seamounts bathymetric data grid (MV1302a) and gas in a submarine crater, a potential (Figure 1a,b), but the new mapping does the two older data sets (2002 EW0202 hazard recently identified at provide a baseline that will help to doc- and 2003 TN153). The resulting map submarine volcano in the Aegean Sea ument future submarine eruptions and (Figure 1e) shows a substantial area of (Carey et al., 2013). slope failures. Differencing of pre- and negative depth change (up to ~ –200 m The ocean acts as a damper for post-event bathymetric surveys has been deeper on post-eruption survey) at and most submarine eruptions. Increased very effective in studying recent volcanic west of the summit of the northern peak hydrostatic pressure at depth reduces gas eruptions along mid-ocean ridges, on (labeled P1 on Figure 1d,f). It also shows exsolution and the relative high density submarine volcanoes over hotspots, a smaller-magnitude but larger area of and viscosity of water (versus air) rapidly and along intra-oceanic arc submarine positive depth change (shallower on reduces the velocity of pyroclastic volcanoes (Rubin et al., 2012). post-eruption survey) on the western material ejected during an eruption. The available pre-eruption multibeam flank of the seamount directly downslope The maximum depth we need to bathymetric surveys of South Sarigan of the breached crater (up to ~ +50 m). consider for submarine eruption hazards Seamount from 2002 (Atlas Hydrosweep The negative depth change area includes depends partly on the type of hazard, DS-2 multibeam, cruise EW0202) and the crater and a zone extending onto the and presumably is directly related to the 2003 (Simrad EM300 multibeam, cruise upper flank. The deficit on the upper eruption’s Volcanic Explosivity Index TN153) were incomplete, leaving a data flank is probably the result of mass (see Discussion section below). Until gap over the shallow main peak (marked wasting likely induced by loading and/ recently, this depth was thought to be as P2 on Figure 1c). A single multibeam or triggering associated with ~ 500 m (Mastin and Witter, 2000) to swath (EW0202) barely covered the the 2010 eruption seismicity episode. generate explosive and tsunami hazards, north peak (marked as P1 on Figure 1c), We interpret the positive depth change at least for all but the very largest the purported location of the 2010 area as a large mass of material added submarine eruptions. However, a recent eruption (McGimsey et al., 2010; Searcy, to the slope during the 2010 eruptions.

Oceanography | June 2014 27 Depth change volumes were calculated of the wall and on the crater’s rim, an South Sarigan magmas are intermediate with QPS Fledermaus © software. Using extensive white filamentous microbial in silica content (Figure 2h), dominated the MV1302a/TN135 depth change grid, mat covered boulders and outcrops by -bearing two a volume for the downslope deposit (Figure 2e). This microbial mat is (Tamura et al., 2013). These (there is not complete coverage so this probably sustained by a diffuse (barely andesites are found as in the is only an estimate) is ~ 1.7 x 108 m3. discernible on the video) hydrothermal cores and as lava blocks on the deposit Calculating the volume lost from the flow. There were no observations of surface and on crater-wall outcrops. summit area by differencing a grid that vent-endemic macrofauna communities Andesitic pumice (which could be from combines the two older surveys with and no indicators of active high-​ the eruption) and lava blocks (which are the MV1302a grid yields a volume of temperature venting. The crater floor probably older lavas flows) show similar 7 3 ~ 6.8 x 10 m . This makes the deposit and wall seen on this traverse did not limited ranges of SiO2 (54–56 wt % and volume about 2.5 times the volume of “look” young—everything was coated 54–58 wt %, respectively) and MgO the summit deficit. As discussed further with dull brown/orange material, contents (4.0–6.5 wt % and 3.5–6.0 wt %, below, the greater volume of the flank and sediments were thicker on the respectively; Figure 2h). K2O contents of deposit relative to the summit deficit benches and on the crater floor. No the lava blocks ranges from 0.2–0.8 wt %, is consistent with the deposit being a young-looking lava flows were observed, whereas pumices contain 0.6–0.8 wt % combination of slope failure and juvenile although this dive traversed very little of K2O. Incompatible trace element ratios volcanic products. the crater floor. However, one indicator further highlight the narrower range that the crater has relatively “young” of pumice compositions (Figure 2h) OBSERVATIONS AND SAMPLING seafloor is the lack of sessile macrofauna compared to the lavas. Pumices have a AT THE ERUPTION SITE on the outcrops, such as the deep corals much smaller range of Ba/Th (300–350) Following the discovery of the summit commonly found on the hard substrata and La/Sm (2.0–2.2) than do lava blocks crater and downslope deposit, two dives of the shallow seamounts of the Mariana (Ba/Th = 250–700 and La/Sm =1.4–2.5, with the Japan Agency for Marine-Earth arc (Rogers, 2007). respectively). These values are similar Science and Technology (JAMSTEC) Dive HPD-1534 traversed the middle to those of primitive from Pagan remotely operated vehicle (ROV) of the thickest portion of the downslope Volcano in the Mariana arc to the north Hyper-Dolphin explored and sampled deposit, from 1,540 to 1,355 m water (Figure 1a,b, inset) reported by Tamura both the crater and the western deposit depth (Figure 2f). The seafloor was et al. (2014). Because of the abundance in June 2013 (Tamura et al., 2013; strewn with large, jointed-appearing, of andesite and the absence of other Figure 2a). Dive HPD-1533 (Figure 2b) partially buried andesite blocks embed- lithologies in the suite collected in the began at 409 m water depth on a sandy, ded in sandy pumiceous volcaniclastic crater (HPD-1533) and in the downslope boulder-strewn seafloor near the base deposits (Figure 2g). Three push cores deposit (HPD-1534), we conclude that of a steep scarp; the ROV then moved taken at ~ 0.5 km intervals along the the recent eruptions of South Sarigan— north to northeast up the wall to the ~ 1.0 km traverse recovered pumiceous including that of May 2010—have been crater rim at ~ 239 m water depth. The gravel that was crudely size sorted away dominated by andesite. northeast wall of the crater is faced by a from the summit. The deposit’s surface From the observations and samples series of jointed andesite flows and dikes and the blocks were covered with a thin made during the dives in June 2013, (Tamura et al., 2013; Figure 2c). There orange-brown deposit or precipitate. we can also conclude the following: was white veining and/or streaking Several factors control the explosivity (1) the of the crater is consistent commonly on the outcrops (Figure 2d). of volcanic eruptions, but water- and with it having formed during the 2010 This could be elemental sulfur, which silica-rich magmas tend to erupt more event, (2) some of the jointed andesite can appear light-colored under water violently than water- and silica-poor lavas from the disruption/collapse of due to absorption of yellow wavelengths magmas. Arc magmas in general are the summit are present on the surface of in over short distances. Several water rich, ~ 4% H2O in basaltic magmas the downslope deposit, (3) the location small sediment-covered benches break (Plank et al., 2013), and we assume that (within the downslope positive depth the slope of the crater wall. Near the top this is also the case for South Sarigan. change zone), size grading, and distinct

28 Oceanography | Vol. 27, No. 2 chemistry of the pumiceous deposit a meters 50 HPD1533 kilometers 2 Geog. proj. Geog. proj. b are consistent with it being an eruptive R11 16°37'N -260 product of the 2010 event, (4) the het- HP1534 R10 R07 -240 R08 R09 erogeneous chemistry of the lava blocks -300 -1200 HP1533 -1000 -800 -400

-1400 R06 C03 indicate that they represent more 16°36'N -600 -340 R05

-200 -400 than one lava flow/eruption episode, C02 R04 R03 R02 consistent with them being part of the -400 16°35'N C01 former summit, and (5) some latent heat R01 145°44'E 145°45'E 145°46'E 145°47'E 145°48'E from the eruption/intrusion was still being diffusely discharged at the summit c d in June 2013. The ubiquitous orange coating on the surface of the deposit is enigmatic, but could be either iron oxides precipitated (possibly by micro- bial activity) during the cooling of the deposit or the last layer of (altered?) ash from the water column plume generated during the eruption. e meters 200 HPD1534 f Geog. proj. -1500 DISCUSSION AND -1600 R01 R02 R03 R07 IMPLICATIONS FOR -1400 C01 R04 C02 R05 R06 R09 SUBMARINE VOLCANIC R08 C03

HAZARDS R10 -1300 The results from bathymetry remapping and seafloor observations are broadly 0.8 lava blocks h consistent with the interpretation of the 6 pumice, 2010 0.6 seismic and hydroacoustic data (Green g 5 et al., 2013; Searcy, 2013). During a 0.4 O 2

4 precursory period beginning on May 27, 0.2 Wt % MgO Wt % K eruptions and intrusions produced 54 56 58 54 56 58 Wt % SiO2 Wt % SiO2 700 2.6 water-borne acoustic T-phases and Ba/Th La/Sm volcanic tremor as pressure built up 2.2 500 beneath the summit. The underwater explosion(s) on May 29, 2010, detected 1.8 300

8,000 km away off western Canada 1.4 54 56 58 54 56 58 (Green et al., 2013), marked the parox- Figure 2. Results from June 2013 dives with JAMSTEC Hyper-Dolphin remotely operated vehicle (ROV) ysmal eruption. The roughly two-day at South Sarigan Seamount. The photos were taken with a SEAMAX digital still camera (DSC) mounted eruption period probably produced on the ROV. (a) Bathymetric map (200 m contours) of South Sarigan Seamount showing locations of dives HPD-1533 and HPD-1534. (b) Dive track of HPD-1533 on the summit crater wall with collection the breached summit crater along with of rock and core samples designated “R” and “C.” (c) Photo of andesite outcrop on the north wall of the juvenile pyroclastic and hydroclastic crater (358 m water depth). (d) SEAMAX DSC photo shows white staining on fractures in the crater wall deposits. Volume calculations derived (278 m water depth) that is possibly elemental sulfur. (e) SEAMAX DSC photo shows the microbial mat that covers boulders near the top/rim of the crater (240 m water depth). (f) Dive track of HPD-1534 on from the bathymetric difference downslope deposit with rock and core sampling indicated as in Figure 2b. The bathymetry underlay has a grids show that the deposit contains 20 m contour interval. (g) SEAMAX DSC photo of a large columnar-jointed andesite block (1,435 m water ~ 2.5 times the volume of that removed depth). (h) Variation of MgO and K2O (top boxes) and Ba/Th and La/Sm ratios (bottom boxes) with SiO2 (wt %) for downslope pumice recovered from cores (yellow squares) and andesites from downslope from the summit. This observation and blocks and summit crater wall (gray squares). Pumices are from dives HPD-1534 and andesite blocks are the relative chemical homogeneity of the from HPD-1533 and HPD-1534.

Oceanography | June 2014 29 downslope pumice fragments compared Green et al. (2013). submarine volcanism. At present, these to the lava blocks imply that at least The South Sarigan event is one of the are limited to just a few sites in the some of the young flank deposits rep- first instances of an explosive, relatively Indian, Pacific, and Atlantic . resent juvenile erupted material. These deep, submarine eruption that breached The existing arrays can only detect some analyses also indicate that the large the surface ocean and for which we have submarine eruptions, and the locations jointed blocks observed on the surface of quantitative data for the size and extent are not always robust, particularly in the downslope deposit originated from of the cratering event and deposits to complex areas. This was exemplified by the summit. However, much more of the match with seismic and hydroacoustic a submarine eruption in the northern summit material is presumably buried monitoring information. Submarine Mariana arc that occurred while this in the deposits (up to ~ 50–75 m thick craters the size of the one formed paper was being finalized (May 2014). in places) and/or lie upslope from the during the eruption of South Sarigan are The locations of the seismic events dive area. The downslope deposits are relatively common on seamounts along using the far-field hydroacoustic arrays volcaniclastic that likely include juvenile intra-oceanic arcs (Figure 1a,b shows were not precise enough to definitively material, such as pumice, and fragmental those along the Mariana arc). This event, locate the eruption site, although, at material (including the large blocks) and a deeper and much larger event at least initially, it is thought to be Ahyi produced by disintegration and collapse Havre Seamount in the Kermadec arc in (http://www.volcano.si.edu/volcano. of the summit. 2012 (Carey et al., 2014), underscores cfm?vn=284141), a seamount lying How does the South Sarigan eruption how little is known of the eruption southeast of the island of Farallon de compare to arc andesite eruptions on history of most submarine arc volcanoes. Pajaros (Figure 1a). Clearly, we need land? The only commonly used index Many Mariana arc submarine edifices to work toward a monitoring system for the size of volcanic eruptions is are complex composite volcanoes with that will more effectively cover the the Volcanic Explosivity Index (VEI). multiple summits. Eruptions of the size most potentially hazardous submarine However, most parameters used as of the South Sarigan, although not his- volcanoes. There is much work to do if criteria for the VEI (e.g., height of torically well documented, are probably we are to better understand the hazards eruption cloud) don’t directly apply to common on a geologic time scale. How associated with submarine arc volcanoes. the South Sarigan event because the many of these seamounts are dormant indices were developed for a solid Earth/ versus extinct and how do we recognize ACKNOWLEDGMENTS atmosphere interface rather than a solid the difference? Most intra-oceanic arc, We are grateful to the NOAA Vents Earth/ocean/atmosphere eruption. , and other potentially active Program (now Earth Oceans Interaction At this point, we know that: (1) there submarine volcanoes have only minimal Program) and the NOAA Office of was a short (~ 2 days) eruptive period sampling of dredged rocks of undeter- and Research for culminating with a large explosive event, mined age. There is clearly a need not long-term funding support of the (2) a large crater (~ 350 m diameter) only to sample these sites but also to Mariana arc explorations. JAMSTEC formed sometime during the eruption, conduct in situ studies of the slopes and provided funding for Y. Tamura, the and (3) the deposit volume was 2.5 times summits in order to reconstruct their Hyper-Dolphin ROV dives, and the larger than the volume of the crater and geologic histories. support vessel Natsushima. This work material removed by summit collapse. Monitoring submarine volcanoes was supported in part by the JSPS The explosive nature of the eruption and is also a prescient issue. At present, Grant-in-Aid for Scientific Research the estimated juvenile deposit volume of monitoring of submarine volcanic [(B) (23340166)]. The 2002 and 2013 ~ 1 x 108 m3 places the eruption within activity is challenging because the global mapping cruises were supported by the moderate-large to large (VEI = 3–4) seismic arrays are mostly limited to the National Science Foundation (2013 category, roughly comparable to the coverage of continental areas and usually cruise OCE-0841074). We thank the destructive Mt. Pelée eruption (Siebert don’t record far-field volcanic seismicity. crews and science teams on the research et al., 2010). The size of the deposit For the foreseeable future, we will vessels T.G. Thompson, Ewing, Melville, is also consistent with that inferred probably depend on hydroacoustic and and Natsushima for their hard work at from the hydroacoustic records by island seismometer arrays to monitor sea. We are grateful to J. White, M. Perfit,

30 Oceanography | Vol. 27, No. 2 G. McGimsey, and J. Chaytor for their Journal of Volcanology and Geothermal Plank, T., K.A. Kelley, M.M. Zimmer, E.H. Hauri, Research 257:31–43, http://dx.doi.org/​10.1016/​ and P.J. Wallace. 2013. Why do mafic arc mag- helpful reviews and C. Searcy and j.jvolgeores.2013.03.006. mas contain ~ 4 wt% water on average? Earth D.N. Greene for informative discussions Hovland, M., and O.T. Gudmestad. 2001. Potential and Planetary Science Letters 364:168–179, about the history of seismic and hydro- influence of gas hydrates on installa- http://dx.doi.org/10.1016/j.epsl.2012.11.044. tions. In Natural Gas Hydrates: Occurrence, Rogers, A.D., A. Baco, H. Griffiths, T. Hart, acoustic monitoring in the southern Distribution, and Detection. C.K. Paull and and J.M. Hall-Spencer, eds. 2007. Corals Mariana arc. PMEL contribution W.P. Dillon, eds, Geophysical Monograph on seamounts. Pp. 141–169 in Seamounts: Series, American Geophysical Union, Ecology, Fisheries & Conservation. T.J. Pitcher, number 4139. Washington, DC, http://dx.doi.org/10.1029/ T. Morato, P. J.B. Hart, M.R. Clark, N. Haggan, GM124p0307. and R.S. Santos, eds, Blackwell Publishing, REFERENCES Kano, K. 2003. Subaqueous pumice eruptions Singapore. from their products: A review. In Explosive Rubin, K.H., S.A. Soule, W.W. Chadwick Jr., Beget, J.E. 2000. Volcanic tsunamis. Subaqueous Volcanism. J.D.L. White, D.J. Fornari, D.A. Clague, R.W. Embley, Pp. 1,005–1,014 in Encyclopedia of Volcanology. J.L. Smellie, and D.A. Clague, eds, Geophysical E.T. Baker, M.R. Perfit, D.W. Caress, and H. Sigurdsson, ed., Academic Press, San Diego. Monograph Series, American Geophysical R.P. Dziak. 2012. Volcanic eruptions in the Bloomer, S.H., R.J. Stern, and N.C. Smoot. Union, Washington, DC, http://dx.doi.org/​ . Oceanography 25(1):142–157, 1989. Physical volcanology of the submarine 10.1029/140GM14. http://dx.doi.org/​10.5670/oceanog.2012.12. Mariana and volcano arcs. Bulletin of Latter, J.H. 1981. Tsunamis of volcanic Searcy, C. 2013. Seismicity associated with the Volcanology 51:210–224, http://dx.doi.org/​ origin: Summary of causes, with particular May 2010 eruption of South Sarigan Seamount, 10.1007/BF01067957. reference to Krakatoa, 1883. Bulletin of northern Mariana Islands. Seismological Carey, R., R. Wysoczanski, R. Wunderman, Volcanology 44:467–490, http://dx.doi.org/​ Research Letters 84(6):1,055–1,061, and M. Jutzeler. 2014. Discovery of the 10.1007/BF02600578. http://dx.doi.org/10.1785/0220120168. largest historic silicic submarine eruption. Maeno, F., F. Inamura, and H. Taniguchi. Siebert, L., T. Simkin, and P. Kimberly. 2010. Eos Transactions, American Geophysical 2006. Numerical simulation of tsunamis Volcanoes of the World, 3rd ed. University of Union 95(10):157–159, http://dx.doi.org/​ generated by caldera collapse during the California Press, Berkeley, CA. 10.1002/2014EO190001. 7.3 ka Kikai eruption, Kyushu, Japan. Tamura, Y., R.W. Embley, A.R. Nichols, O. Ishizuka, Carey, S., P. Nomikou, K.C. Bell, M. Lilley, Earth, Planets, and Space 58:1,013–1,024, S.G. Merle, W.W. Chadwick Jr., R.J. Stern, J. Lupton, C. Roman, E. Stathopoulou, http://www.terrapub.co.jp/journals/EPS/ T. Sato, D.A. Wiens, and P. Shore. 2013. K. Bejelou, and R. Ballard. 2013. CO degassing 2 abstract/5808/58081013.html. ROV Hyper-Dolphin survey at the May 2010 from hydrothermal vents at Kolumbo subma- Mastin, L.G., and J.B. Witter. 2000. The eruption site on South Sarigan Seamount, rine volcano, Greece, and the accumulation of hazards of eruptions through lakes and Mariana Arc. Paper presented at the 2013 Fall acidic crater water. Geology 41:1,035–1,038, seawater. Journal of Volcanology and Geothermal Meeting of the American Geophysical Union, http://dx.doi.org/10.1130/G34286.1. Research 97:195–214, http://dx.doi.org/10.1016/ San Francisco, California, Abstract V31G-02. Dietz, R.S., and M.J. Sheehy. 1954. Transpacific S0377-0273(99)00174-2. Tamura, Y., O. Ishizuka, R.J. Stern, A.R. Nichols, detection of Myojin volcano explosions McCoy, F.W., and G. Heiken. 2000. Tsunami H. Kawabata, Y. Hirahara, Q. Chang, by underwater sound. Geological generated by the Late Bronze Age eruption T. Miyazaki, J.-I. Kimura, R.W. Embley, Society of America Bulletin 65:941–956, at Thera (Santorini), Greece. Pure and and Y. Tatsumi. 2014. Mission immiscible: http://dx.doi.org/10.1130/0016-7606​ Applied Geophysics 157:1,227–1,256, Distinct subduction components generate two (1954)65[941:TDOMVE]2.0.CO;2. http://dx.doi.org/10.1007/s000240050024. primary magmas at Pagan Volcano. Journal of Embley, R.W., E.T. Baker, D.A. Butterfield, McGimsey, R.G., C.A. Neal, C.K. Searcy, Petrology 55:63–101, http://dx.doi.org/10.1093/ W.W. Chadwick Jr., J.E. Lupton, J.A. Resing, J.T. Camacho, W.B. Aydlett, R.W. Embley, petrology/egt061. C.E.J. de Ronde, K. Nakamura, J.F. Dower, and F.A. Trusdell, J.F. Paskievitch, and Vaughan, R.G., M.J. Abrams, S.J. Hook, and S.G. Merle. 2007. Exploring the submarine D.J. Schneider. 2010. The May 2010 submarine D.C. Pieri. 2007. Satellite observations of new : Mariana Arc, western Pacific. eruption from South Sarigan Seamount, volcanic island in . Eos Transactions, Oceanography 20(4):68–79, http://dx.doi.org/​ Northern Mariana Islands. Paper presented at American Geophysical Union 88(4):37–41, 10.5670/oceanog.2007.07. 2010 Fall Meeting of the American Geophysical http://dx.doi.org/10.1029/2007EO040002. Embley, R.W., C.E.J. de Ronde, and J. Ishibashi. Union, December 13–17, Abstract T11E-07. Watt, S.F.L., P.J. Talling, and J.E. Hunt. 2014. 2008. Introduction to special section on active Niino, H. 1952. Explosions of Myojin . New insights into the emplacement magmatic, tectonic, and hydrothermal pro- Kagaku Asahi December:3–23. dynamics of volcanic island landslides. cesses at intraoceanic arc submarine volcanoes. Nomanbhoy, N., and K. Satake. 1995. Oceanography 27(2):46–57, http://dx.doi.org/​ Journal of Geophysical Research 113, 808S01, Generation mechanism of tsunamis 10.5670/oceanog.2014.39. http://dx.doi.org/10.1029/2008JB005871. from the 1883 Krakatau eruption. Esposito, A., G. Giordano, and M. Anzidei. 2006. Geophysical Research Letters 22:509–512, The 2002–2003 submarine gas eruptions http://dx.doi.org/10.1029/94GL03219. at Panarea volcano (, Paris, R., A.A. Switzer, M. Belousova, A. Belousov, Italy): Volcanology of the seafloor and B. Ontowirjo, P.L. Whelley, and M. Ultvrova. implications for the hazard scenario. Marine 2013. Volcanic tsunami: A review of source Geology 227:119–134, http://dx.doi.org/​ mechanisms, past events and hazards in 10.1016/j.margeo.2005.11.007. Southeast Asia (Indonesia, Philippines, Papua Green, D.N., L.G. Evers, D. Free, R.S. Matoza, New Guinea). Natural Hazards 70:447–470, M. Snellen, P. Smets, and D. Simons. http://dx.doi.org/10.1007/s11069-013-0822-8. 2013. Hydroacoustic, infrasonic and seismic monitoring of the submarine eruptive activity and sub-aerial plume generation at South Sarigan, May 2010.

Oceanography | June 2014 31